Soft robotics is a branch of robotics that deals with mechatronics and electromechanical systems primarily made of soft materials. This paper presents a summary of a chronicle study of various soft robotic hand exoskeletons, with different electroencephalography (EEG)- and electromyography (EMG)-based instrumentations and controls, for rehabilitation and assistance in activities of daily living. A total of 45 soft robotic hand exoskeletons are reviewed. The study follows two methodological frameworks: a systematic review and a chronological review of the exoskeletons. The first approach summarizes the designs of different soft robotic hand exoskeletons based on their mechanical, electrical and functional attributes, including the degree of freedom, number of fingers, force transmission, actuation mode and control strategy. The second approach discusses the technological trend of soft robotic hand exoskeletons in the past decade. The timeline analysis demonstrates the transformation of the exoskeletons from rigid ferrous materials to soft elastomeric materials. It uncovers recent research, development and integration of their mechanical and electrical components. It also approximates the future of the soft robotic hand exoskeletons and some of their crucial design attributes.

Sensory data are critical for soft robot perception. However, integrating sensors to soft robots remains challenging due to their inherent softness. An alternative approach is indirect sensing through an estimation scheme, which uses robot dynamics and available measurements to estimate variables that would have been measured by sensors. Nevertheless, developing an adequately effective estimation scheme for soft robots is not straightforward. First, it requires a mathematical model; modeling of soft robots is analytically demanding due to their complex dynamics. Second, it should perform multimodal sensing for both internal and external variables, with minimal sensors, and finally, it must be robust against sensor faults. In this article, we propose a recurrent neural network-based adaptive unscented Kalman filter (RNN-AUKF) architecture to estimate the proprioceptive state and exteroceptive unknown input of a pneumatic-based soft finger. To address the challenge in modeling soft robots, we adopt a data-driven approach using RNNs. Then, we interconnect the AUKF with an unknown input estimator to perform multimodal sensing using a single embedded flex sensor. We also prove mathematically that the estimation error is bounded with respect to sensor degradation (noise and drift). Experimental results show that the RNN-AUKF achieves a better overall performance in terms of accuracy and robustness against the benchmark method. The proposed scheme is also extended to a multifinger soft gripper and is robust against out-of-distribution sensor dynamics. The outcomes of this research have immense potentials in realizing a robust multimodal indirect sensing in soft robots.

There has been a boost of research activities in robotics using soft materials in the past ten years. It is expected that the use and control of soft materials can help realize robotic systems that are safer, cheaper, and more adaptable than the level that the conventional rigid-material robots can achieve. Contrary to a number of existing review and position papers on soft-material robotics, which mostly present case studies and/or discuss trends and challenges, the review focuses on the fundamentals of the research field. First, it gives a definition of soft-material robotics and introduces its history, which dates back to the late 1970s. Second, it provides characterization of soft-materials, actuators and sensing elements. Third, it presents two general approaches to mathematical modelling of kinematics of soft-material robots; that is, piecewise constant curvature approximation and variable curvature approach, as well as their related statics and dynamics. Fourth, it summarizes control methods that have been used for soft-material robots and other continuum robots in both model-based fashion and model-free fashion. Lastly, applications or potential usage of soft-material robots are described related to wearable robots, medical robots, grasping and manipulation.

Soft sensors are widely used in many industrial systems to monitor key variables that are difficult to measure, using measurements from other available physical sensors. Because physical sensors are susceptible to faults, it is crucial for soft sensor models to be robust against them. Recently, deep learning has shown promising results in developing data-driven soft sensors for various applications. However, existing learning-based soft sensors are still vulnerable to sensor faults, which could deteriorate the performance of the models. In this article, we propose a deep learning-based modeling framework for developing soft sensor models that are robust to sensor faults. Due to the difficulty in obtaining datasets that cover all possible sensor fault characteristics, the proposed framework is developed to be zero-shot such that the model can be trained with only fault-free dataset without requiring any sensor fault patterns, thus greatly saving the time and resources needed to collect such data. Instead, adversarial examples are used as a proxy for faulty sensor inputs so that the model can learn to be adaptive through the proposed two-stage, uncertainty-aware recurrent neural network architecture. We demonstrate our approach to the TE benchmark process and a real industrial multiphase flow process and show that robustness is achieved as the accuracy does not degrade significantly when sensor faults are present during the model evaluation.

The mechanical compliance of soft robots comes at a cost of higher uncertainty in their sensing and perception, which deteriorates the accuracy of predictive models. Predictive uncertainty, which expresses the confidence behind model predictions, is necessary to compensate for the loss of accuracy in soft robot perceptive models. Nevertheless, developing a general framework to capture uncertainties is further challenged by the complex dynamics of soft robots and the difficulties in sensorizing them. In this work, we present a predictive uncertainty estimation framework based on deep learning for soft robot multimodal sensing. We show that the framework can learn to quantify uncertainty and thus is able to express the confidence associated with the predictions during inference. Being data-driven, it is scalable to different types of soft robots and sensor modalities. We demonstrate the framework on a complex multimodal sensing task where a single flex sensor is used to predict the full-body configuration of a soft actuator, as well as the magnitude and location of external contact force. We also discuss how predictive uncertainties are critical to achieve safe learning and model interpretability in soft robotics.

Aerial grasping empowers unmanned aerial vehicles to find applications beyond structured logistics. However, it brings a number of challenges including inaccurate positioning of the end effector and limited energy sources. Moreover, solutions so far have difficulty in handling a variety of objects. A novel closed structure compliant gripper was developed to address the challenges above. The gripper has a large self-centering work envelope and is normally-closed for passive grasping. Introduction of compliance as a form of morphological computation was also considered to enhance grasping capabilities, where multi-material 3D printing would facilitate rapid design changes based on target application. To grasp different objects, the gripper has hot-swappable 3D printed fingertips which are optimized with a multi-objective Bayesian Optimization process using physical bench experiments mimicking drone grasping and ascent on a common object set. The morphology of the fingertip including tip width, curvature and distribution of soft and hard material on contact surface, are optimized with a bench test that mimics quadcopter takeoff and landing. The best design from optimization shows an improvement of more than 10% from the initial design in successful grasp operations, demonstrated by field tests with a quadcopter.